User personalization for the web is an effective tool for gaining insights into website traffic and improving the user experience. Traditional methods involve carefully analysing the user environment and identifying useful parameters to track. However, this process is heavily dependent on human expertise and can be challenging to automate in an ever-changing environment. This thesis proposes a novel approach by utilising reinforcement learning to automate the personalization process and provide the system with the ability to adapt to changes in the user environment. The primary objective of this thesis is to evaluate the potential benefits of using reinforcement learning in automating the web personalization process as an alternative approach to enhance the efficiency and effectiveness of user personalization for the web for both users and website owners. The goal is to investigate the use of reinforcement learning as a non-traditional method in automating the web personalization process and its effectiveness in providing a tailored and personalized user experience. In addition, this thesiswill investigate a way to recreate synthetic data, and develop a simulation pipeline to evaluate the performance of reinforcement learning based personalization methods under different scenarios. The aim of this thesis is to determine the feasibility of using reinforcement learning for automating the web personalization process as a solution.